One of the major difficulties in studying acquired immunodeficiency syndrome (AIDS) is the lack of an adequate experimental system. The SCID-hu mouse represents a murine model in which interactive organs of the human hematolymphoid system (fetal liver, thymus, lymph node) have been engrafted into an immunodeficient mouse strain (C,B17 scid/scid), in such a fashion that human progenitor cell differentiation may be observed to proceed. When HIV is introduced into these organs by direct injection, infection is found to spread in a time-dependent and dose- dependent manner; such events may also be observed after the intravenous inoculation of virus. the goal of this project is to develop the SCID-hu model so that it might be used to study the pathogenesis of HIV infection and for the design of more effective therapeutic regimens. Basic to this goal are the requirements that the therapeutic regimens. Basic to this goal are the requirements that the system be standardized, reproducible, quantifiable, and safe to use. The titers of various isolates of HIV (most of them derived from molecular clones) will be normalized by standard assays in vitro; infection of the SCID-hu mouse will then permit direct correlation of the "infectious unit" in vivo. Standard inputs of virus will be delivered by various routes and in various forms, to create a reproducible system for infection. Thereafter, the histopathologic effects of infection and the human immune response to infection will be examined. Finally, reproducible and quantitative assays for infection will be developed. These assay will be designed to permit the safe evaluation of anti-virals and/or prophylactic agents in a statistically meaningful way.